Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Mar 10.
Published in final edited form as: Glia. 2017 Mar 16;65(6):945–963. doi: 10.1002/glia.23137

HuR Promotes the Molecular Signature and Phenotype of Activated Microglia: Implications for Amyotrophic Lateral Sclerosis and Other Neurodegenerative Diseases

Prachi Matsye 1,5, Lei Zheng 1,5, Ying Si 1,5, Soojin Kim 1, Wenyi Luo 6, David K Crossman 2, Preston E Bratcher 4, Peter H King 1,2,3,5,*
PMCID: PMC7944581  NIHMSID: NIHMS1672569  PMID: 28300326

Abstract

In neurodegenerative disease like amyotrophic lateral sclerosis (ALS), chronic activation of microglia contributes to disease progression. Activated microglia produce cytokines, chemokines and other factors that normally serve to clear infection or damaged tissue either directly or through the recruitment of other immune cells. The molecular program driving this phenotype is classically linked to the transcription factor NF-κB and characterized by the upregulation of pro-inflammatory factors such as IL-1β, TNF-α, and IL-6. Here, we investigated the role of HuR, an RNA binding protein that regulates gene expression through posttranscriptional pathways, on the molecular and cellular phenotypes of activated microglia. We performed RNA sequencing of HuR-silenced microglia and found significant attenuation of lipopolysaccharide-induced IL-1β and TNF-α inflammatory pathways and other factors that promote microglial migration and invasion. RNA kinetics and luciferase reporter studies suggested that the attenuation was related to altered promoter activity rather than a change in RNA stability. HuR-silenced microglia showed reduced migration, invasion and chemotactic properties but maintained viability. MMP-12, a target exquisitely sensitive to HuR knockdown, participates in the migration/invasion phenotype. HuR is abundantly detected in the cytoplasmic compartment of activated microglia from ALS spinal cords consistent with its increased activity. Microglia from ALS-associated mutant SOD1 mice demonstrated higher migration/invasion properties which can be blocked with HuR inhibition. These findings underscore an important role for HuR in sculpting the molecular signature and phenotype of activated microglia, and as a possible therapeutic target in ALS and other neurodegenerative diseases.

Introduction

Microglia are versatile cells of the central nervous system (CNS) and play diverse but sometimes opposing roles in healthy and diseased brain. These functions include synaptic plasticity, neuronal circuitry, immune surveillance, neuroinflammation, inflammatory suppression and neuroprotection (David and Kroner 2011; Kettenmann et al. 2011; Ransohoff et al. 2015; Wake et al. 2013). A dynamic and broad molecular repertoire allows microglial cells to adopt these different roles, both morphologically and biochemically, depending on cues from the microenvironment. Exposure to infectious agents or abnormal proteins, for example, may activate microglia (classically referred to as the M1 phenotype), whereby they assume an amoeboid morphology, migrate toward the stimulus and secrete inflammatory cytokines such as IL-1β, TNF-α and/or IL-6 (Block and Hong 2005; Hanisch 2002; Kettenmann et al. 2011). On the other hand, release of BDNF in response to other environmental signals may modulate neuronal excitability or promote neuronal survival (Block and Hong 2005; Glass et al. 2010; Wake et al. 2013).

The role of microglia in neuroinflammation has gained much interest in recent years as mounting evidence has linked these cells to disease progression in amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS), Alzheimer’s disease (AD) and Parkinson’s disease (PD) (Cunningham 2013; Henkel et al. 2009; Nguyen et al. 2002). In neuroinflammation, activated microglia in the vicinity of diseased or injured CNS tissue become the “first responders” and sound the alarm by producing cytokines and chemokines that amplify the inflammatory response through recruitment and activation of other immune cells (Hanisch 2002). The molecular driver of the M1 phenotype has classically been attributed to NF-κB activation and subsequent transcription of a program of pro-inflammatory genes (Glass et al. 2010; Nguyen et al. 2002). However, gene expression is a multi-step process, with posttranscriptional regulation playing an essential role in mRNA processing, translocation to the cytoplasm and translation into protein (Anderson 2010). Many of the factors that drive the pro-inflammatory response, such as IL-1β, IL-6, COX-2, and TNF-α, are heavily regulated posttranscriptionally through adenine- and uridine-rich elements (ARE) in the 5’ and 3’ untranslated regions (UTRs) of the mRNA (Abdelmohsen et al. 2008; Barreau et al. 2006; Chen and Shyu 1995). These elements govern mRNA stability and translational efficiency through interactions with RNA binding proteins. HuR is one such protein that contains three RNA recognition motifs (Brennan and Steitz 2001; Srikantan and Gorospe 2012). Several of the early phenotype assessments of HuR were done in models of colon and brain cancer where it was found to be overexpressed and linked to upregulation of tumor-promoting genes containing AREs in the 3’ UTR (Dixon et al. 2001; Nabors et al. 2001). Many of the targets upregulated in these tumors overlap with the inflammatory gene profile in activated microglia. In the current study, we hypothesized that HuR plays an important role in the regulation of genes associated with activated microglia. This hypothesis was bolstered by our findings (shown in this report) that HuR is upregulated in activated microglia in mouse and human ALS spinal cords, and localized prominently in the cytoplasm, a pattern associated with its increased activity. To gain insight into the impact of HuR on gene expression profiles, we performed RNA sequencing on activated microglia in which HuR was knocked down. We observed attenuation of a large number of genes associated with inflammation, chemotaxis, and migration. With HuR knockdown, there was a significant loss of microglial cell mobility, invasion, and the capacity for these cells to recruit other immune cells including microglia and neutrophils. We show that MMP-12, a target which had the largest attenuation with HuR knockdown, participates in microglial cell migration. With HuR inhibition, we were able to suppress inflammatory cytokine induction and migration/invasion properties of ALS-associated mutant SOD1 microglia, suggesting that posttranscriptional pathways could be a therapeutic target in diseases driven by neuroinflammation.

Materials and Methods

Mice and cell culture

C57BL/6J mice purchased from Jackson Lab (Sacramento, CA) were housed in pathogen free environment according to the National Institute of Health and the University of Alabama at Birmingham Institutional Animal Care and Use Committee guidelines. Primary microglial cells (PMG) were isolated from 1–3 day old pups as described previously (Giulian and Baker, 1986). In short, after chemical dissociation of the dissected brains, the mixed glial cultures were maintained in the incubator at 37°C with humidified 5% CO2/ 95% air in T75 flasks (Thermo Fisher Scientific, Waltham, Massachusetts) in DMEM-F12 medium (Corning, Corning, NY) with 20% FBS (Sigma-Aldrich, St. Louis, MO). Fresh medium supplemented with GM-CSF (Pepro Tech, Rocky Hill, NJ) at 10 ng/ml was added the next day and half of the medium was changed after 7 days. PMG were harvested 14 days after isolation by shaking at 195 rpm for 45 mins at 37°C. For subsequent experiments, FBS was reduced to 10%. The purity of PMG was assessed by Iba1 staining. BV2 cells were a gift from Dr. Tika Benveniste (University of Birmingham at Alabama) and were cultured as described (Petrova et al. 1999).

Next generation RNA Sequencing and analysis

mRNA sequencing was performed on the Illumina HiSeq2500 (Illumina Inc., Sand Diego, CA). Total RNA from PMG was purified and assessed for quality using the Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA). Only samples with a RNA Integrity Number (RIN) of 7.0 or above were used, and sequencing libraries were prepared with the SureSelect Strand Specific mRNA kit as per the manufacturer’s instructions (Agilent). For library construction, the polyA mRNA was randomly fragmented, followed by first strand synthesis using random primers with inclusion of Actinomycin D (2.4 ng/μL final concentration). Second strand cDNA production was done with standard techniques and the ends of the resulting cDNA were made blunt and A-tailed. Adaptors were ligated for indexing to allow for multiplexing during sequencing. The cDNA libraries were quantitated using qPCR in a Roche LightCycler 480 (Roche Diagnostics, Indianapolis, IN) with the Kapa Biosystems kit for Illumina library quantitation (Kapa Biosystems, Woburn, MA) prior to cluster generation. Cluster generation was performed according to the manufacturer’s recommendations with for board clustering (Illumina). Paired end 50 bp sequencing runs were completed to allow for better alignment of the sequences to the reference genome. For data assessment, TopHat version 2.0.12 was used to align the raw RNA-Seq fastq reads to the mouse mm10 genome using the short read aligner Bowtie version 2.2.3 (Langmead et al. 2009; Trapnell et al. 2009; Trapnell et al. 2012). Cufflinks version 2.2.1 was used to align reads from TopHat to assemble transcripts, estimate their abundances and test for differential expression and regulation (Trapnell et al. 2012; Trapnell et al. 2010). Assembled transcripts were merged to a reference annotation with Cuffmerge. Cuffdiff was used to search for significant changes in transcript expression, splicing and promoter use. Genes that met certain criteria (i.e. greater than 2-fold and q-value < 0.05) were further analyzed using Ingenuity’s Pathway Analysis tool (www.ingenuity.com). For generating networks, a data set containing gene identifiers and corresponding expression values was uploaded into Ingenuity’s Pathway Analysis tool (Qiagen, Redwood City, CA). Each identifier was mapped to its corresponding object in Ingenuity’s Knowledge Base. A fold change cutoff of ± 2.0 and p-value < 0.05 was set to identify molecules whose expression was significantly differentially regulated. These molecules, called Network Eligible molecules, were overlaid onto a global molecular network developed from information contained in Ingenuity’s Knowledge Base. Networks of Network Eligible Molecules were then algorithmically generated based on their connectivity. The Functional Analysis identified the biological functions and/or diseases that were most significant to the entire data set. Molecules from the dataset that met the fold-change cutoff of ± 2.0 and p-value < 0.05 and were associated with biological functions and/or diseases in Ingenuity’s Knowledge Base were considered for the analysis. Right-tailed Fisher’s exact test was used to calculate a p-value determining the probability that each biological function and/or disease assigned to that data set is due to chance alone.

Plasmids, inhibitors, siRNAs and transfections

The Trem1 promoter luciferase plasmid was kindly provided by Dr. Nagaoka (Juntendo University, Tokyo, Japan) (Hosoda, 2011). The CIAP2 promoter luciferase plasmid was provided by Dr. Katijima (Dept. of Biochemical Genetics, Tokyo Medical and Dental University) (Hua et al. 2006).The pGL4.32[luc2P/NF-κB-RE/Hygro] luciferase reporter plasmid contianing 5 copies of the NF-κB response element-was obtained from Promega (Madison, WI). SMARTpool ON-TARGETplus siRNA for HuR and the control siRNA (GFP) were purchased from GE Dharmacon (Lafayette, CO). BV2 cells were transfected by electroporation using NeonTM transfection system (Invitrogen, Carlsbad, CA) according to the manufacturer’s guidelines. PMG cells were transfected in 6-well plates using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Cells were plated at a density of 5 × 105 cells and 300 pmoles of siRNA were used per transfection. For both cell types, 5 × 105 were used per transfection and seeded in 6-well plates. All transfections were performed in triplicate. After transfection, cells were grown for 48 h (BV2) or 72 h (PMG) before stimulating with LPS (1 μg/ml). For inhibitor studies, MS-444 (HuR inhibitor) was provided by Dr. Nicole Meisner (Novartis) and MMP408 (MMP-12 inhibitor) was purchased from EMD Millipore. For promoter studies, luciferase reporter plasmids (8 μg), siGFP or siHuR, and SV40-β-galactosidase plasmid (2 μg) were co-transfected. Luciferase activity was assayed 24hrs after LPS stimulation using a kit (Promega) and measured with a Synergy 2 multimode microplate reader (Bio-Tek, Winooski, VT) and normalized to β-galactosidase activity as previously described (Nabors et al. 2003).

Western blot, antibodies, ELISA, and immunohistochemistry

Whole cell extracts were prepared using M-Per kit (Thermo Fisher Scientific, Waltham, Massachusetts) and quantitated with a bicinchoninic acid protein (BCA) assay kit (Thermo Fisher Scientific, Waltham, Massachusetts). Lysates (30 μg) were resolved by electrophoresis using 4–20% gels (Bio-Rad, Philadelphia, PA) and blotted on nitrocellulose membrane. The membrane was then probed with primary antibody for HuR (1:1000) (Santa Cruz, Santa Cruz, CA) for 24 h, followed by the secondary antibody (1:3000) and developed using SuperSignal West Dura Extended Duration Substrate kit (Thermo Fisher). The membrane was stripped and re-probed with a GAPDH primary antibody (Cell Signaling, Boston, MA). Densitometric analysis was performed using Lab Imaging software (Bio-Rad, Philadelphia, PA). ELISAs for IL1β, IL6, CXCL1 (R&D systems, Minneapolis, MN) and MMP-12 (Boster, Pleasanton, CA) of conditioned media from BV2 and PMG cells were performed according to manufacturer’s instructions. For immunohistochemistry, 10 μm frozen sections of human or mouse spinal cord were fixed in Bouin’s fixative for 15 min and then oxidized in 0.3% H2O2 for 15 min. Human tissue was obtained post-mortem from patients in the UAB ALS clinic as part of a protocol approved by the UAB Institutional Review Board. Control tissue was provided by the UAB Tissue Procurement Facility. After blocking, sections were incubated with anti-Iba1 1:200 (Wako, Cambridge, MA) and HuR (1:200) (Santa Cruz, Santa Cruz, CA) antibodies overnight at 4°C. The next day, slides were incubated for two hours at RT with donkey anti-rabbit 488 and donkey anti-mouse cy3 secondary antibodies (Jackson ImmunoResearch, West Grove, PA). The prepared slides were viewed under TCS SP5 Visible-Upright Confocal Microscope (Leica Microsystems, Buffalo Grove, IL). Quantitative analysis of HuR/Iba1 colocalization was performed on 4 random low power views of lumbar cord sections from mutant or wild-type SOD1 mice using ImageJ software (National Institutes of Health, Bethesda, Maryland) and the Green and Red Puncta colocalization macro for ImageJ (Daniel J. Shiwarski, Ruben K. Dagda and Charleen T. Chu). Cells meeting the threshold for merged signal were expressed as a percent of total Iba1-positive cells in the field.

RNA isolation, cDNA preparation, qRT-PCR and kinetics

Total RNA was extracted using GE Healthcare RNAspin Mini kit and quantified using NanoDrop2000 (Thermo Fisher Scientific, Waltham, Massachusetts). RNA (1–2μg) was reverse transcribed according to manufacturer’s instructions (Thermo Fisher Scientific, Waltham, Massachusetts). Multiplex quantitative RT-PCR was performed as described previously (Nabors et al. 2003) using ViiA7 Real-Time PCR system (Thermo Fisher Scientific, Waltham, Massachusetts). Commercially available mouse Taqman primers and probes were purchased from Applied Biosystems. Target mRNA signal values were normalized to GAPDH in the same reaction. For mRNA kinetics, after siHuR treatment for 24 or 48 h, cells were treated with Actinomycin D for up to 6 h, and RNA was isolated at different time points. RNA levels were quantitated by qRT-PCR and degradation curves were generated with GraphPad (GraphPad Software).

Cross-Linking Immunoprecipitation Assay

Cross-Linking Immunoprecipitation (CLIP) was performed as described elsewhere (Bose et al. 2006). Briefly, BV-2 cells (8×107) were harvested by centrifugation at 100g for 5 min at 4°C and then suspended in 10 ml of PBS. Formaldehyde (Sigma-Aldrich) was added to the cell suspension to a final concentration of 0.5% (v/v) and the reaction was incubated at room temperature for 10 min with slow mixing. Cross-linking was quenched by the addition of glycine (pH 7.0, 0.25 M final concentration), followed by incubation at room temperature for 5 min. The cells were harvested by centrifugation, followed by 2 washes with ice-cold PBS. Fixed cells were resuspended in 1 ml of RIPA buffer [50 mM Tris/HCl (pH 7.5), 1% NP-40, 0.05% SDS, 1 mM EDTA and 150 mM NaCl] containing protease inhibitors. The cells were lysed by 3 rounds of sonication for 20 s each. An aliquot of cell lysate was mixed with 20μl of Pierce Protein A/G Agarose beads (Pierce) for 1 h at 4°C followed by centrifugation at 400g for 5 min. The pre-cleared supernatant was diluted with an equal volume of RIPA buffer containing RNasin and protease inhibitors, mixed with equivalent amounts of anti-HuR IgG or IgG isotype control antibodies (Santa Cruz) and incubated over-night with shaking at 4°C. Pierce Protein A/G Agarose beads (Pierce) (100μl) were added and the samples were incubated with shaking for overnight at 4◦C. The beads were washed five times with RIPA buffer and then resuspended beads were incubated at 70°C for 45 min to reverse the cross-linking. RNA was extracted from the immunoprecipitates using the GE Illustra RNA spin kit. The RNA was reverse transcribed and qPCR was performed using mouse Taqman primers and probes. Calculations for fold-enrichment of HuR IgG were based on methods detailed in the Imprint® RNA Immunoprecipitation Kit Manual (Sigma-Aldrich). Briefly, Ct values for IgG and anti-HuR IgG were normalized to the input RNA using the following equation: ΔCt [normalized CLIP] = (Ct [CLIP] − (Ct [Input] −Log2 (Input Dilution Factor)). The fold-enrichment of anti-HuR IgG was calculated as 2−ΔΔCt where ΔΔCt = ΔCt [normalized HuR IgG] −ΔCt [normalized IgG].

Cell Viability and Phagocytosis

Cell viability assay was performed using a kit (Lonza, Allendale, NJ) according to manufacturer’s instructions. Phagocytosis assay was performed using FluoSpheres Carboxylate-Modified Microspheres (Invitrogen, Carlsbad, CA). Prior to assay, microspheres were washed overnight in PBS with 1%BSA with gentle shaking at RT. Microglia (2.5 × 104 cells) were allowed to settle overnight in 2-chamber slides (Thermo Fisher Scientific). Five hundred beads per cell were added to the wells and incubated for 1 h. Cells were then washed thoroughly with PBS and fixed in 4% paraformaldehyde for 10 min and then stained with DAPI.

Cell Chemotaxis, Migration, and Invasion Assays

Murine bone marrow neutrophils were isolated from the tibia and femur of C57/BL6J mice using a Percoll gradient method described elsewhere (Taylor et al. 2014). Purity of cells was monitored by flow cytometry where neutrophils were defined as Ly6G+ cells. Using 3.0 μm pore size 96-well MIC plates (EMD Millipore, Billerica, MA), 2 × 105 cells were added to the top chamber in 100 μl of RPMI medium, and 150 μl of the indicated conditioned medium was placed into the bottom chamber. Medium supplemented with CXCL1 (R&D systems, Minneapolis, MN) at 800 ng/ml was used as a positive control. Plates were incubated in a humid incubator at 37° C and 5% CO2 for 1 h, after which the top chambers were removed and cells in the bottom chamber were counted using a hemocytometer. The assay was repeated twice with three biological and two experimental replicates each time. For BV2 and PMG chemotaxis, 2 × 104 cells in 100 μl of serum-free media were loaded on the top chamber of a 8.0 μm pore size 24-well transwell migration plate (Corning, Corning, NY) and 500 μl conditioned media was placed in the bottom chamber. The plates were incubated at 37°C and 5% CO2 for 24 h. The media was then aspirated and cells on the inside of the insert were wiped off thoroughly using a Q-tip. The inserts were then cut out and stained using the Hema 3 Stat Pack (Thermo Fisher Scientific, Waltham, Massachusetts) for 2 min each stain. Inserts were allowed to dry and then mounted on the slides with oil and coverslip. Experiments were done in triplicate and 20 fields under high power objective were captured per replicate. Cell migration experiments were performed in a similar manner using treated cells with serum-free media in the top chamber and media with 300 μM ATP in the bottom chamber. For cell invasion assays, 8.0 μm pore size 24-well matrigel invasion chamber plates (Corning, Corning, NY) were used and the assay was performed as described above. For the inhibitor studies, MS-444, MMP408, or vehicle control (DMSO) was added to the culture medium during the 24 h migration period.

Results

HuR modulates expression of key mediators of inflammation and cell migration in primary microglial cells

Mouse primary microglia (PMG) were isolated and transfected with siRNA directed to HuR. Compared to control siRNA, we observed a drop in HuR expression by 80% (Fig.1A). After knockdown, cells were stimulated with LPS for 24 h and total RNA was analyzed by next-generation RNA sequencing. When filtered for targets with a false discovery rate (q value) < 0.05 and at least a 2-fold change, 172 differentially expressed targets were identified, with 86% being attenuated upon HuR knockdown (Fig. 1 and Supp. Info. Table1). While these targets can reflect indirect effects (e.g. related to modulating the expression of a transcription factor), the large percentage of suppressed targets is consistent with HuR’s general role as a positive regulator of gene expression. We utilized the Ingenuity Pathway Analysis tool to assess potential biological functions of the 172 mRNA targets (Fig. 1B). A number of cellular functions were significantly linked to the targets including inflammatory response, cell movement, and proliferation, all of which are features of activated PMG (Kettenmann et al. 2011). Other categories related to cell movement, including cellular infiltration/invasion and migration were also significantly represented. Interestingly a number of the cell movement associations in the analysis are based on data from tumor cells. The findings are consistent with what we have observed in vivo with glioblastoma multiforme, a highly malignant brain tumor, where HuR silencing significantly reduces tumor invasion and tumor cell migration (Filippova et al. 2011). Two major inflammatory pathways in activated PMG, IL-1β and TNF-α, were predicted to be significantly inhibited with HuR knockdown (Fig. 1C). Since IL-1β mRNA itself was reduced (2.6-fold), some of the downstream targets may represent an indirect effect of HuR silencing. Further inspection of individual targets attenuated by HuR silencing provides support for a role in promoting neuroinflammation (Table 1). mRNAs encoding ligands that regulate chemotaxis of other cells (Cxcl1, Cxcl2, Cxcl3, and Csf3) and receptors that respond to chemotactic signals (Ccr2, Ccr 3, Ccr 5, Nrp1) were significantly reduced. Genes involved in extracellular matrix (ECM) breakdown, a process necessary for cellular migration, were also attenuated, such as plasminogen activator urokinase (Plau), thrombospondin 1, matrix metalloproteinases (MMP)-12 and 13, integrin beta 3, thrombospondin 1, and podoplanin. Multiple components of inflammatory signaling were affected including IL-6, IL-1α, IL-1β, and Ptgs (Cox-2), and two amplifiers of cytokine production, Trem-1 and TremL4. We previously observed that Trem1 is regulated in astrocytes by KH-type splicing regulatory protein (KSRP), another ARE binding protein which promotes degradation and reduced translation of its target mRNAs (King and Chen 2014; Li et al. 2012). Of the classic factors associated with the M2 phenotype, arginase 1 was significantly attenuated by nearly 4-fold (Supp. Info. Table 2). IL-10 and TGF-β1, on the other hand, were not significantly different. Other M2-associated factors such as IL-4, 13 and YM1 were not detected, and thus no conclusions can be drawn regarding HuR regulation. Because HuR can regulate translational efficiency of the mRNA, at times separate from RNA stability, there are limitations to interpreting the number of targets affected by HuR knockdown (Srikantan and Gorospe 2012). Some targets may only show changes in protein expression and would not be identified in the RNA sequencing analysis. Overall, however, the data indicate that HuR modulates targets in PMG cells that control aspects of the activated phenotype.

Figure 1: HuR modulates genes involved in diverse biological functions in activated PMG.

Figure 1:

Open in a new tab

(A) Western blot showing knockdown of HuR in PMG after transfection with HuR siRNA. To the right of the blot, results of quantitative densitometry showing percent knockdown of HuR compared to control siRNA. The pie chart shows a summary of affected targets in HuR-silenced microglia that had a false discovery rate (q value) of < 0.05 and a 2-fold or greater change compared to control siRNA treated cells. (B) Biological functions significantly represented based on analysis of affected gene targets using the Ingenuity Pathway Analysis tool (see Materials and Methods). P values are shown on the X axis. (C) IL-1β and TNF-α pathways were predicted to be significantly inhibited with HuR knockdown (P < 2.0 × 10−9). The different intensity shades of red/green indicate the intensity of fold change (darker red, more up-regulated; darker green, more down-regulated). Ingenuity colored the relationship lines based off the known literature and the fold-change values from our dataset.

Table 1.

siHuR-affected RNA targets in primary microglia that are relevant to inflammation/immune function and cell migration1

Symbol ARE2 Fn3 Gene Name Δ
Olr1 + I Oxidized Low Density Lipoprotein (Lectin-Like) Receptor 1 −10.9
Mmp12 + M,I Matrix Metallopeptidase 12 −6.5
Thbs1 + M Thrombospondin 1 −6.3
Cxcl3 + I, C Chemokine (C-X-C Motif) Ligand 3 −6.3
Itgb3 + M Integrin, Beta 3 −5.9
Gdf15 + I Growth differentiation factor 15 −4.6
Clec7a + I C-Type Lectin Domain Family 7, Member A −4.3
Ccr2 + I, C Chemokine (C-C Motif) Receptor 2 −4.1
Cd28 + I Cluster of differentiation 28 −4.1
Il23r − I Interleukin 23 Receptor −3.7
Il7r − I Interleukin 7 Receptor −3.6
Txk − I Tyrosine Kinase −3.5
Kitl + M kit ligand −3.3
Il1a + I, C Interleukin 1, Alpha −3.1
Havcr2 − I Hepatitis A Virus Cellular Receptor 2 −2.9
Pla2g5 − I Phospholipase A2, Group V −2.8
Nrp1 − M Neuropilin 1 −2.8
Cxcl1 + I, C Chemokine (C-X-C Motif) Ligand 1 −2.8
Cxcl2 + I, C Chemokine (C-X-C Motif) Ligand 2 −2.8
Ptgs2 + I Prostaglandin-Endoperoxide Synthase 2 (COX2) −2.7
Il1b + I, C Interleukin 1, Beta −2.6
Ccr5 + I, C Chemokine (C-C Motif) Receptor 5 −2.6
Ccr3 + I, C Chemokine (C-C Motif) Receptor 3 −2.5
Trem1 − I Triggering Receptor Expressed On Myeloid Cells 1 −2.5
Mmp13 + M Matrix Metallopeptidase 13 (Collagenase 3) −2.5
CD32 − I FcγII receptor −2.5
Plau + M, C Plasminogen Activator, Urokinase −2.4
Csf3 + I Colony Stimulating Factor 3 (Granulocyte) −2.4
Pdpn − M podoplanin −2.3
Treml4 − I Triggering Receptor Expressed On Myeloid Cells-Like 4 −2.3
Baiap2 + M BAI1-Associated Protein 2 −2.3
Il6 + I, C Interleukin 6 (Interferon, Beta 2) −2.2
Adamts1 + M, C ADAM Metallopeptidase With Thrombosp. Type 1 Motif, 1 −2.2
Csf2rb2 − I Colony Stimulating Factor 2 Receptor, Beta, 2 −2.1
Open in a new tab
1

Only targets with > 2-fold change compared to control with Q values (false discovery rate) < 0.05

2

ARE, Au-rich element in the 3’ untranslated region

3

Related to: C, chemotaxis; I, inflammation/immune responses; M, migration and invasion

Validation of targets identified by RNA sequencing in PMG and BV2 microglial cells

We next sought to validate a subset of the targets identified by RNA sequencing, both at the mRNA and protein level, and to determine if a similar pattern occurred in BV2 microglial cells after HuR knockdown. BV2 cells transfected with siHuR showed an ~80% reduction in HuR compared to control siRNA (Fig. 2A). Overall we observed significant declines in all RNA targets in PMG cells, a validation of the RNA seq findings. BV2 cells showed a similar pattern after LPS, with some differences in degree of change. MMP-12 and 13 were 4–6-fold lower in BV2 versus PMG, whereas Trem-1 was more than 2-fold lower in PMG. TNF-α mRNA, which was unchanged in the RNA sequencing analysis of PMG, was also not affected by HuR knockdown in BV2 cells. In other cell systems, HuR has been linked to the stability and upregulation of TNF-α mRNA (Dean et al. 2001; Katsanou et al. 2005). We tested iNOS, an important component of microglial activation and neuroinflammation, which has been linked to HuR in other cell systems (Rodriguez-Pascual et al. 2000). This target was attenuated in our RNA seq analysis (p = 0.03, q = 0.28; not shown), and the reduction was validated in both PMG and BV2 cells. We next looked at the production of secreted proteins (Fig. 2C). We used ELISA for protein detection in conditioned media from PMG or BV2 cells after stimulation with LPS. Comparisons were made to cells transfected with siRNA control. We observed significant reductions of IL-1β, IL-6, CXCL1, and MMP-12, with the latter target being diminished by more than 5-fold. Interestingly, TNF-α production was significantly reduced in PMG and trending downward in BV2 cells. With the mRNA levels unaffected, this result suggests that HuR is modulating translational efficiency of this target rather than mRNA expression level. In summary, the RNA sequencing data were validated in both PMG and BV2 cells.

Figure 2: Validation of targets in PMG and BV-2 microglial cells.

Figure 2:

Open in a new tab

A subset of targets identified by RNA sequencing was further assessed in PMG and BV2 cells after HuR knockdown. (A) Western blot and quantitative densitometry of BV-2 cells showing knockdown of HuR after siRNA transfection as compared to the control siRNA. (B) Quantitation of different targets by qRT-PCR. Values for each target were first adjusted to the internal GAPDH control and then expressed as a relative quantity (RQ) to the mean value of three independent siRNA control samples (set at 1.0). All data points are the mean ± SD of three independent samples. (C) Protein expression of some secreted targets as determined by ELISA of conditioned media. Values for control and HuR siRNA treated cells are the mean ± SD of three independent samples. The mean control siRNA value was set at 1.0. * P < 0.05.

HuR binds to mRNA targets but knockdown attenuates promoter activity rather than mRNA stability

We first determined whether HuR bound to the target mRNAs using a cross-linking immunoprecipitation assay with LPS-stimulated BV2 cells (Fig. 3). We observed binding for all the targets of interest, with IL-6 and CXCL1 showing the highest fold-increase over IgG control. Treatment of the cells with MS-444, a small molecule which inhibits HuR homodimerization and RNA binding (Meisner et al. 2007) suppressed binding of all the targets by 2–3 fold. We next determined whether decreases in mRNA targets following HuR knockdown were related to changes in mRNA stability. HuR was silenced in BV2 cells with siRNA for 48 h followed by LPS stimulation. Transcription was inhibited with Actinomycin D (ActD) and the target mRNA was quantified at different time points and compared to baseline RNA (time 0). Half-lives were estimated at the time point when 50% of the RNA remained. Most of the mRNA targets were fully stabilized (half-lives exceeding the 4–6 h time course), with siHuR and control siRNA showing similar patterns (Fig. 4). These very same targets showed significant decreases with HuR knockdown (e.g. MMP-12). Only for Cxcl1 was the mRNA destabilized with HuR knocked down (half-life ~2.4 h vs. > 4 h for control). TNF-α, a target mRNA which was not altered with HuR knockdown, showed no change in its typically short half-life. To ensure that a change in half-life was not transient early after HuR knockdown, the experiment was repeated 24 h after siRNA transfection (Supp. Fig. 1). Again, no destabilization was observed after HuR knockdown for any of the targets, including Cxcl1, while IL-6 showed a longer half-life than control siRNA. These findings suggest that RNA stability is not playing a key role in the down regulation of many target mRNAs following HuR knockdown. We cannot exclude an artifactual effect of ActD on the stabilization of these targets which has previously been observed with other ARE bearing mRNAs, and has been known to alter the subcellular localization of HuR in some cell types (Dean et al. 2001; Peng et al. 1998; Shyu et al. 1989).

Figure 3: HuR binds to target mRNAs and binding is attenuated by the HuR inhibitor, MS-444.

Figure 3:

Open in a new tab

BV2 cells were stimulated with LPS and subjected to RNA immunoprecipitation with anti-HuR IgG or control IgG. Some cells were treated concomitantly with MS-444 at 40 μM. Immunoprecipitated mRNA was quantified using qRT-PCR and adjusted to the input RNA as described in Materials and Methods. Quantities of mRNA pulled down by HuR IgG are expressed as a fold-change over IgG control. Data points represent the mean ± SD of two independent samples. The experiment was repeated one time with similar results.

Figure 4: Knockdown of HuR does not alter RNA stability in microglial cells for most targets.

Figure 4:

Open in a new tab

We assessed the RNA half-life of several targets in BV2 microglial cells after HuR knockdown. Following siRNA transfection and LPS stimulation, cells were pulsed with actinomycin D and RNA was collected at different time intervals. The target mRNAs were quantified by qRT-PCR and compared to pre-actinomycin levels as a percentage. Half-lives were estimated as the time point when 50% of the RNA remained (shown in parentheses). All data points represent the mean ± SD of three independent samples.

Our findings raise the possibility that HuR is affecting transcription of these mRNAs. To address this possibility, we used a luciferase reporter system with promoters from two targets, IL-1β and Trem-1, a control promoter, CIAP2, whose mRNA . level was not affected by HuR knockdown, and an NF-kB response element reporter (Fig.5A). Plasmids were transfected into BV2 cells and then stimulated with LPS for 24 h. We observed significant suppression of luciferase activity with the IL-1β and Trem1 promoters, but not CIAP2 (Fig. 4A). A similar pattern of promoter suppression was observed in PMG (Supp. Info. Fig. 2). The 5X NF-κB minimal promoter showed a 40% reduction in activity raising the possibility that some of the suppressive effect is related to inhibition of NF-κB activity. In a previous report, we observed that overexpression of the RNA destabilizer KSRP, in astrocytes, suppressed promoter activity of Trem1 and TNF-α with possible involvement of an NF-kB site (Li et al. 2012). Interestingly, knockout of KSRP did not affect the basal half-lives of mRNA targets that were significantly increased. In summary, these findings raise the possibility that with some targets, the effect of HuR knockdown is through altered promoter activity.

Figure 5: HuR knockdown suppresses promoter activity of several targets.

Figure 5:

Open in a new tab

Left panel: schematic diagram of promoter constructs used in the analysis. Right panel: luciferase activity of the different promoters following transfection into BV2 cells where HuR was knocked down by siRNA. Luciferase values for the control siRNA were set at 1. All data points represent the mean ± SEM of 3–4 independent samples. *P < 0.05.

Knockdown of HuR negatively affects microglial chemotaxis and migration

Since HuR modulates factors that have a direct impact on microglial migration and chemotaxis of immune cells, we used a transwell assay to assess conditioned medium from siHuR-treated microglial cells. Untreated BV2 or PMG cells were added to the upper well and conditioned medium from siHuR-treated BV2 cells was used as a chemoattractant in the lower well. After 24 h, inserts were stained and migrated cells were counted. There was a 50% reduction in BV2 cells and an 80% reduction in PMG cells compared to siRNA control (Fig. 6A). Bone marrow derived murine neutrophils were assessed for chemotaxis in a similar approach by placing cells in the upper well. We observed a reduction of neutrophil chemotaxis by 50% with BV2 medium and 30% with PMG medium (Fig. 6B). These results indicate that HuR promotes the inflammatory phenotype of microglial cells by augmenting production of factors that enhance chemotaxis of neutrophils and other microglial cells. We next assessed microglial migration. For these assays, siHuR-treated BV2 or PMG cells were stimulated with LPS for 24 h, and then seeded in the upper well. Growth medium supplemented with ATP was added in the lower chamber as a chemoattractant. BV2 cell migration was reduced by 50% compared to control siRNA (Fig. 7A). Since the capacity for PMG migration in the CNS depends on breakdown of the ECM, we used a transwell with matrigel coated on the upper well to assess invasion properties of PMG. There was a remarkable 4-fold reduction in migrated PMG following HuR knockdown, suggesting a defect in invasion. To ensure that these migration deficits were not related to cytotoxicity, we used a luciferase-based assay to detect ATP as an indicator of cell viability or toxicity (Slater 2001). Following siRNA treatment and LPS stimulation, lysates were prepared and luminescence was measured after the addition of luciferin and luciferase. We observed no decrease in luminescence in siHuR-treated cells as compared to control, indicating that there was no change in viability (Fig. 7B). Phagocytic function was also not affected by HuR knockdown as determined by a fluorescent microsphere uptake assay (Supp. Info. Fig. 3). As an alternative approach to suppressing HuR, we treated BV2 and PMG cells with MS-444 and observed a significant and dose-dependent decrease in migration similar to when HuR was knocked down (Fig. 8). In summary, loss of HuR significantly impaired microglial migration and chemotaxis.

Figure 6: HuR knockdown in microglial cells attenuates chemotaxis.

Figure 6:

Open in a new tab

(A) HuR was knocked down in BV2 cells and conditioned medium was collected after stimulation with LPS. Untreated PMG or BV2 cells were placed in the upper well of a transwell migration plate and conditioned medium in the bottom well. After 24 h, migrated cells were counted as described in Materials and Methods. (B) Neutrophil migration was tested in a similar way using conditioned medium from HuR-silenced BV2 or PMG cells in the bottom well. The data points are the mean ± SD of two independent experiments each with three biological replicates. * P < 0.05.

Figure 7: HuR knockdown attenuates migration and invasion of microglial cells.

Figure 7:

Open in a new tab

(A) BV2 or PMG cells were treated with siHuR RNA or control siRNA and placed in the upper well of a transwell migration plate (with matrigel for PMG). Medium containing ATP was placed in the bottom well as a chemoattractant. After 24 h, migrated cells were counted. All data points represent the mean ± SD of three independent experiments. *P < 0.05. (B) Viability of microglial cells (BV2, left panel; PMG, right panel) following HuR knockdown and LPS treatment was assessed by a luciferase-based assay. This assay indirectly measures ATP production as a measure of cell viability. RLU, relative light units.

Figure 8: MS-444, a small molecule inhibitor of HuR, attenuates microglial migration and invasion.

Figure 8:

Open in a new tab

BV2 (left panel) and PMG (right panel) cells were plated in the upper chamber of a transwell plate and treated with MS-444 at the doses shown. ATP in the lower chamber was used as a chemoattractant. At 24 h, migrated cells were stained and counted as shown. Data points are the mean ± SD of 3 independent samples (counting cells in 20 random high power fields for each replicate). * P < 0.05.

MMP-12 contributes to invasion/migration properties of microglial cells

Based on the marked attenuation of MMP-12 levels with HuR knockdown (Fig. 2), coupled with its essential role in macrophages for tissue invasion (Shipley et al. 1996), we hypothesized that this metalloproteinase participates in the migration of PMG. We assessed this potential role by using the chemical inhibitor MMP408 (Fig. 9A). We observed a dose-dependent decrease in BV2 migration upon treatment with the inhibitor, with 70% reduction at higher doses (P < 0.05). This finding was confirmed in PMG which showed a 50% reduced invasion at a similar dose. Similarly, knockdown of MMP-12 in BV2 cells with siRNA significantly reduced migration (Fig. 9B). Taken together, these results show that MMP-12 contributes to the migration phenotype of microglial cells.

Figure 9: Microglial cell migration is attenuated by MMP-12 suppression.

Figure 9:

Open in a new tab

(A) BV2 cells (left panel) and PMG (right panel) were plated in the upper chamber and the MMP-12 inhibitor was added. ATP was used as a chemoattractant in the lower chamber. (B) BV2 cells were transfected with MMP-12 siRNA, treated with LPS, and plated in the upper chamber. ATP was added to the lower chamber as the chemoattractant. Cells were counted and values were expressed as a percent of the mean siRNA control value (± SD). (C) ELISA of MMP-12 in the conditioned media from MMP-12 siRNA-transfected BV2 cells in the experiment shown in (B). Data points for all cell counting are the mean ± SD of 3 independent samples (counting cells in 20 random high power fields for each replicate). * P < 0.05.

HuR is upregulated in ALS-associated mutant SOD1 microglia and inhibition attenuates cytokine induction, cell migration and invasion

Neuroinflammation is a common feature of ALS and is characterized by extensive microglial activation and infiltration of peripheral immune cells at sites of neurodegeneration (Alexianu et al. 2001; Hall et al. 1998; Kawamata et al. 1992; Philips and Rothstein 2014; Turner et al. 2004). Mutant SOD1 (mtSOD1) microglia exhibit increased expression of many pro-inflammatory genes, and we next determined the impact of HuR on that molecular phenotype using the target genes studied in this report (Chiu et al. 2013; Philips and Robberecht 2011; Weydt et al. 2004; Zhao et al. 2013). We first assessed HuR expression in ALS-associated mutant SOD1 (mtSOD1) PMG by Western blot and found it to be increased compared to WT (Fig. 10A). RNA expression levels of HuR-modulated mRNA targets were all significantly higher in mtSOD1 PMG following LPS stimulation (Fig. 10B). The absence of a change in TNF-α was similar to a prior report of LPS-stimulated neonatal microglia (Weydt et al. 2004). With HuR knockdown, cytokine induction was significantly suppressed to levels below control except for IL-6 which trended downward. TNF-α again was unchanged (Fig. 10C). We next determined the impact of HuR inhibition on PMG migration using the transwell model and ATP as a chemoattractant. At baseline, mtSOD1 PMG demonstrated a two-fold increase in migrated cells versus wild-type cells (Fig. 10D). Treatment of the cells with MS-444 led to a dose-dependent decrease in migration/invasion, with essentially no migrated cells being detected at 20 μM (Fig. 10E). Taken together, these results indicate that HuR and its mRNA targets are upregulated in mtSOD1 PMG and that HuR silencing or inhibition can attenuate these targets and the properties of invasion and migration.

Figure 10: HuR inhibition attenuates elements of the activated phenotype in mutant SOD1 PMG.

Figure 10:

Open in a new tab

(A) Western blot detection of HuR in LPS-stimulated PMG from mtSOD1 or WT mice. (B) Comparison of RNA targets in LPS-stimulated WT and mtSOD1 microglia by qPCR. The mean WT value was set at 1.0 after adjustment for the internal GAPDH control. (C) Comparison of RNA targets in siHuR or siCtl-treated mtSOD1 microglia induced with LPS. (D) Transwell assay measuring invasion/migration of WT or mtSOD1microglia after 24 h. ATP was used as a chemoattractant. (E) Dose-dependent effect of MS-444 on mtSOD1 microglial invasion/migration. All Values represent the mean ± SD of three independent samples.

HuR is increased in the cytoplasmic compartment of activated microglia in ALS spinal cords

We used immunohistochemistry to detect HuR expression in microglia from spinal cords of mtSOD1 and WT mice. In this mouse model, diffuse infiltration of activated microglia is detected in the spinal cord by day 120, a time which correlates with onset of weakness and significant motor neuron loss (Alexianu et al. 2001; Hall et al. 1998). In the lumbar spinal cord of a 125 day old mtSOD1 mouse, we observed an abundant number of cells with intense Iba1 immunoreactivity consistent with activated microglia (Fig. 11). These cells also had prominent HuR immunoreactivity, consistent with our finding of increased HuR expression in activated mtSOD1 microglia (Fig. 10). In contrast, age-controlled WT mice showed fewer microglia with more muted Iba1 and HuR immunoreactivity. The number of Iba1 positive cells in mtSOD1 mice exceeded that of WT by 2.5 fold consistent with a prior report (Hall et al. 1998). In most of the mtSOD1 microglia, there was prominent merging of HuR and Iba1 signals, indicating an abundance of HuR in the cytoplasmic compartment. Microglia from WT mice showed some merged signal within the cytoplasm but considerably less compared to mtSOD1. Using a colocalization macro for ImageJ, we quantitated the number of microglia with high intensity merged HuR and Iba1 signal. There was a three-fold increase in positive cells for mtSOD1 versus WT microglia. These findings are consistent with an increase of HuR in the cytoplasm of activated microglia in mtSOD1 spinal cords. We cannot definitively conclude that there is active translocation as the immunofluorescent signal may be the result of overall increased expression of HuR and Iba1 in these disease-associated cells. To determine if a similar pattern was present in humans, we assessed spinal cord sections from three ALS and one control patient. Two of the patients had sporadic ALS and one had familial ALS (A68T mutation in SOD1). The control was an age-equivalent patient who died of cardiac arrest. As with the ALS mouse, we observed multiple cells with prominent merged Iba1 and HuR signal in all three ALS spinal cords, but not the control, consistent with an increased presence of HuR in the cytoplasmic compartment (Fig. 12). Taken together, these findings suggest that HuR is upregulated in activated microglia of ALS spinal cords and abundantly present in the cytoplasm.

Figure 11: HuR colocalizes with Iba1 in activated microglia from mtSOD1 spinal cords.

Figure 11:

Open in a new tab

(A) Representative photomicrographs of HuR and Iba1 immunoreactivity in mtSOD1 or age-matched WT lumbar spinal cords (125 d postnatal). Antibodies are shown at the top. Scale bars, 10 μm. (B) Quantitation of Iba1 positive cells in mtSOD1 or WT lumbar cords (125 d postnatal). (C) Quantitation of cells with merged HuR and Iba1 signal using ImageJ (see Materials and Methods). Data points represent the mean ± SEM of 4 random fields. ***, P < 0.001.

Figure 12: HuR colocalizes with Iba1 in microglia from human ALS spinal cords.

Figure 12:

Open in a new tab

Immunohistochemistry of spinal cord sections from ALS or control patients using HuR and Iba1 antibodies. Antibodies are shown at the top. Scale bars, 10 μm. ALS1, 64-year-old male, sporadic; ALS2, 73-year-old-male, sporadic; ALS3, 40-year-old-male, familial (A68T SOD1 mutation); Ctl, 63-year-old-male, cardiac arrest.

DISCUSSION

In this report, we show that knockdown of the RNA binding protein, HuR, in microglia attenuates production of pro-inflammatory cytokines, chemokines, and factors that modulate cell migration and invasion. We show significant downstream effects including defects in chemotaxis, migration and invasion. Our findings bring into focus posttranscriptional gene regulation as an important level of molecular control that works in concert with classical transcriptional pathways such as NF-κB to establish the molecular signature of activated microglia (Figuera-Losada et al. 2014; Glass et al. 2010). These findings are relevant to ALS where we show significant upregulation of HuR in activated microglia and an increased abundance in the cytoplasmic compartment. Inhibition of HuR in ALS-associated microglia blocked cytokine induction and the phenotypes of chemotaxis and invasion, underscoring the potential therapeutic implications for this disease and other neurodegenerative or neuroinflammatory disorders accelerated by persistent microglial activation.

Messenger RNA production in the nucleus, under the direction of transcription factors such as NF-κB and AP-1, represents only one step of gene regulation in response to microenvironment triggers. In association with RNA binding proteins, these nascent mRNAs undergo maturation, transportation to the cytoplasm, and translation into proteins, the ultimate effector molecules of the regulatory cascade. Posttranscriptional pathways such as RNA stability and translational efficiency can quickly turn on or off gene expression and represent a major control point in immune cells for dampening or accentuating inflammatory/immune responses (Anderson 2008). ARE-mediated posttranscriptional regulation represents a major control pathway for inflammatory cytokines and chemokines as many contain AREs in the 3’ UTR (Anderson 2008). A number of RNA binding proteins bind to the ARE, but promote different fates such as RNA degradation, translation, or storage in granules (Anderson and Kedersha 2006; Barreau et al. 2006). With several exceptions, HuR functions as a positive regulator of RNA stability and translational efficiency (Brennan and Steitz 2001; Srikantan and Gorospe 2012). Other ARE-binding proteins such as KSRP and TTP are negative regulators and promote RNA degradation and translational silencing (Abdelmohsen et al. 2008; Srikantan and Gorospe 2012). Determinants of RNA stability therefore reflect a balance between stabilizing and destabilizing factors, which can be altered by differences in expression levels or post-translational modifications such as phosphorylation or methylation (Barreau et al. 2006; Srikantan and Gorospe 2012). Our RNA sequencing data are supportive of HuR generally as a positive regulator in microglia since the vast majority of RNA targets (with q < 0.05) were attenuated after HuR knockdown (Fig. 1). Many, but not all, of the affected RNA targets are linked to inflammation, chemotaxis and migration and contain AREs in the 3’ UTR amenable to HuR binding (Table 1). In a subset of validated targets, there was no change in RNA half-life after HuR knockdown despite evidence of HuR binding (Figs. 3 and 4). This finding coupled with the luciferase reporter results (Fig. 5) suggests that the attenuated mRNA levels after HuR knockdown were related to suppression of promoter activity. The negative effect on the engineered NF-κB response promoter suggests that this element may be partly involved. One report showed that HuR silencing in endothelial cells suppressed phosphorylation of p65 and other NF- κB signaling components with a concomitant attenuation of mRNAs linked to stress and inflammation (Rhee et al. 2010). We did not observe any effects of HuR knockdown on phosphorylated or total p65 (data not shown). The determinants of NF-κB activity, however, are complex and gene-specific, including NF-κB dimer subtype, affinity to the NF-κB binding sites, susceptibility to binding IκBα, and differential interactions with co-regulatory proteins (Ghosh et al. 2010; Sen and Smale 2010). This complexity is demonstrated in our results showing that the CIAP2 promoter, which contains NF-κB response elements, was not affected by HuR knockdown (Fig. 5). HuR, unlike other RNA binding proteins such as KSRP, TIAR and TDP-43, has not been shown to bind directly to DNA, and we favor the notion of an indirect effect, such as HuR-related changes in expression of a transcription factor, enhancer or suppressor (Buratti and Baralle 2001; King and Chen 2014; Suswam et al. 2005). An analysis of the RNA seq data revealed a number of transcriptional regulators and kinases that could also impact promoter activity (Supp. Table 3). Some of these factors have been linked to NF- κB activity in other cell systems, although sometimes as repressors. The scope of potential targets involved is potentially broader as HuR also modulates translational efficiency. This level of posttranscriptional regulation is distinct and sometimes independent from RNA stability and not always reflected by changes in RNA levels. Our findings with TNF-α are a potential example of this dissociation. Cytochrome C, XIAP, and TDP-43 represent examples in other cell systems (Durie et al. 2011; Kawai et al. 2006; Lu et al. 2014; Srikantan and Gorospe 2012). The attenuated protein levels (Fig. 2) may well be related to this level of regulation as HuR bound to all of the target mRNAs (Fig. 3).

Typically nuclear in location, HuR translocates to the cytoplasm upon activation where it facilitates mRNA association with polysomes or interferes with silencing miRNAs (Abdelmohsen et al. 2008; Brennan and Steitz 2001; Meisner and Filipowicz 2011). The upregulation of HuR in ALS-associated microglia and its increased abundance in the cytoplasm are supportive of a role in promoting the activated phenotype in that disease (Figs. 10–12). The spectrum of microglial activation, however, is broad and ranges from innate to adaptive activation that is governed by stimuli in the microenvironment (Town et al. 2005). Thus, activation may vary based on the diseased or normal states that produce those stimuli. It will be interesting to see whether activation of microglia in other neurodegenerative diseases such as AD or PD shows a similar pattern with HuR.

Based on the biological functions identified by RNA sequencing (Fig. 1 and Table 1), the impact of HuR on the phenotype of activated microglia is potentially broad. One central theme that emerges is the regulatory role HuR plays in the inflammatory response. Knockdown of HuR led to a significant mRNA and/or protein attenuation of key inflammatory markers of microglial activation, including IL-6, IL-1β, TNF-α, Cox-2, IL-1α, and iNOS. The downstream pathways of IL-1β and TNF-α were likewise inhibited (Fig. 1). The effect of HuR on the inflammatory response is further underscored by the positive regulation of Trem1, a transmembrane receptor on myeloid cells that serves to amplify cytokine production in response to inflammatory signals (Bouchon et al. 2001). On the other hand, IL-10 and Tgf-β1, markers associated with an anti-inflammatory phenotype (classically referred to as M2), were not affected at the mRNA level by HuR knockdown (Pena-Altamira et al. 2015). Thus, HuR potentiates expression of many M1 phenotype-associated genes activated through the classic NF-kB transcriptional pathway to promote neuroinflammation.

Another theme that emerges from the RNA seq data is the regulatory role of HuR in migration and chemotaxis (Table 1). The ability to migrate through the ECM is an essential property of microglia for rapid targeting of CNS injury, infection or other perturbations (Inoue 2008; Kettenmann et al. 2011). Breakdown of the ECM is critical for this mobility, and a number of HuR-regulated genes play direct roles in this process, including MMP-12, MMP-13, Adamts1, thrombospondin 1 and urokinase. MMP-12 abundance was exquisitely sensitive to HuR knockdown, ranking near the top for fold-decrease in mRNA. The decline in protein expression was in parallel with a striking 5 to10-fold reduction of secreted protein (Fig. 2). In the CNS, MMP-12 is mainly expressed in microglia and becomes upregulated in pathological states, often by more than 100-fold, including advanced ALS, intracerebral hemorrhage, spinal cord injury, stroke, and acute demyelinating plaques in MS (Chelluboina et al. 2015; Chiu et al. 2013; Crocker et al. 2008; Power et al. 2003; Vos et al. 2003; Wells et al. 2003). While the role of this protein in microglia has not been fully elucidated, it was shown to be essential for macrophage-mediated proteolysis of the ECM and tissue invasion (Shipley et al. 1996). ECM breakdown would also be necessary for proper microglial migration within the CNS, and our findings that blockade of MMP-12, either by chemical inhibition or gene silencing, attenuated migration and invasion in vitro indicates a contributing role of this protein (Fig. 9). Interestingly, microglia from mtSOD1 mice showed more than a two-fold increase in MMP-12 expression and a concomitant two-fold increase in invasion/migration of cells compared to wild-type microglia (Fig. 10). The former is consistent with another report where MMP-12 was increased by 125-fold in microglia from end-stage ALS (Chiu et al. 2013). A therapeutic potential for MMP inhibition in ALS was shown in the mtSOD1 mouse where administration of an MMP inhibitor early in the disease course prolonged survival and improved motor function (Lorenzl et al. 2006). Other deleterious effects linked to MMP-12 include myelin proteolysis, axonal damage and activation/upregulation of other potentially toxic MMPs (e.g. MMP-2, 3, and 9) (Lagente et al. 2009; Liao et al. 2015; Matsumoto et al. 1998; Shiryaev et al. 2009). This may explain the improved outcome of stroke and spinal cord injury in mice where MMP-12 was deleted (Chelluboina et al. 2015; Wells et al. 2003). Other ECM regulators that were affected by HuR knockdown (e.g. MMP-13, Adamts1, podoplanin, or Plau), are also likely contribute to the migration of microglial cells.

As initial responders, microglia must also send out chemotactic signals to recruit additional microglia, astrocytes and immune cells from the periphery (e.g. neutrophils, macrophages or monocytes). Our findings support a role for HuR in microglial recruitment of neutrophils (Fig. 6), and this may stem from the suppression of two major neutrophil chemokines, CXCL1 and 2, (Table 1 and Fig. 2) (Kolaczkowska and Kubes 2013). On the other hand, reduction of CCR2, 3 and 5, three chemotactic response receptors in microglia, could explain the reduced chemotaxis of HuR-silenced microglia (Fig. 6). CCR2 is a major receptor responsible for the accumulation of microglia to diseased areas, such as AD and MS, and its mRNA was suppressed by more than 4-fold with HuR knockdown (El Khoury et al. 2007; Jiang et al. 2012; Mahad and Ransohoff 2003; Semple et al. 2010). CCR3 and CCR5 are also upregulated in activated microglia and found in diseased brain including AD and MS (Simpson et al. 2000; Xia et al. 1998). Additional studies will be required to tease out which affected target is responsible for the observed changes in chemotaxis.

In summary, HuR is a key posttranscriptional regulator of many genes that make up the molecular signature of activated microglia, including increased production of pro-inflammatory cytokines, chemokines and migration-associated factors. Microglia play multiple roles in normal brain and neuroinflammatory/degenerative diseases, some beneficial and some detrimental (Cherry et al. 2014; Pena-Altamira et al. 2015; Streit 2002; Wake et al. 2013). The roles may change depending on the disease type and stage. In the G93A mouse model of ALS for example, microglia are neuroprotective in the early stages of the disease and later become deleterious (Henkel et al. 2009; Zhao et al. 2013). In models of AD and MS, migration of microglia to diseased brain and phagocytic removal of debris (Aβ in the former and myelin breakdown products in the latter) may be beneficial (El Khoury et al. 2007; Neumann et al. 2009). On the other hand, in spinal cord injury, early upregulation of IL-1β and TNF-α (microglia being a major source) contributes to neuronal toxicity (David and Kroner 2011). Taken together, these findings suggest that HuR may be a therapeutic target depending on the disease and stage of disease. HuR expression is not limited to microglia, and thus it remains to be seen whether other cells could be adversely affected by its inhibition.

Supplementary Material

1

Main points.

  • The RNA binding protein HuR modulates expression of genes associated with activated microglia.

  • Loss of HuR attenuates expression of migration-related genes, leading to impaired microglial chemotaxis, migration and invasion.

  • HuR is abundantly detected in the cytoplasmic compartment of activated microglia from human and mouse ALS spinal cords.

  • Mutant SOD1 microglia show enhanced migration and invasion properties which can be suppressed with HuR inhibition.

Acknowledgements

The authors report no conflict of interest. The authors wish to thank the UAB Genomics Core, funded through the UAB Comprehensive Cancer Center (CA13148) and CFAR (AI027767). We would like to thank Dr. Rakesh Bakshi, Department of Medicine (Division of Infectious Diseases) for his kind support with neutrophil isolation and Dr. Ranjit Kumar, Center for Clinical and Translational Science, for his help with RNA sequencing analysis. We are also thankful to our patients who donated their spinal cord tissue post-mortem for ALS research.

References

  1. Abdelmohsen K, Yuki K, Kim HH, Gorospe M. 2008. Posttranscriptional gene regulation by RNA-binding proteins during oxidative stress: implications for cellular senescence. Biol Chem 389:243–255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alexianu ME, Kozovska M, Appel SH. 2001. Immune reactivity in a mouse model of familial ALS correlates with disease progression. Neurology 57:1282–9. [DOI] [PubMed] [Google Scholar]
  3. Anderson P 2008. Post-transcriptional control of cytokine production. Nat Immunol 9:353–9. [DOI] [PubMed] [Google Scholar]
  4. Anderson P 2010. Post-transcriptional regulons coordinate the initiation and resolution of inflammation. Nat Rev Immunol 10:24–35. [DOI] [PubMed] [Google Scholar]
  5. Anderson P, Kedersha N. 2006. RNA granules. J Cell Biol 172:803–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Barreau C, Paillard L, Osborne HB. 2006. AU-rich elements and associated factors: are there unifying principles? Nucleic Acids Res 33:7138–7150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Block ML, Hong J-S. 2005. Microglia and inflammation-mediated neurodegeneration: Multiple triggers with a common mechanism. Prog Neurobiol 76:77–98. [DOI] [PubMed] [Google Scholar]
  8. Bose SK, Sengupta TK, Bandyopadhyay S, Spicer EK. 2006. Identification of Ebp1 as a component of cytoplasmic bcl-2 mRNP (messenger ribonucleoprotein particle) complexes. Biochem J 396:99–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bouchon A, Facchetti F, Weigand MA, Colonna M. 2001. TREM-1 amplifies inflammation and is a crucial mediator of septic shock. Nature 410:1103–1107. [DOI] [PubMed] [Google Scholar]
  10. Brennan CM, Steitz JA. 2001. HuR and mRNA stability. Cell Mol Life Sci 58:266–277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Buratti E, Baralle FE. 2001. Characterization and Functional Implications of the RNA Binding Properties of Nuclear Factor TDP-43, a Novel Splicing Regulator of CFTR Exon 9. J Biol Chem 276:36337–36343. [DOI] [PubMed] [Google Scholar]
  12. Chelluboina B, Warhekar A, Dillard M, Klopfenstein JD, Pinson DM, Wang DZ, Veeravalli KK. 2015. Post-transcriptional inactivation of matrix metalloproteinase-12 after focal cerebral ischemia attenuates brain damage. Sci Rep 5:9504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen CY, Shyu AB. 1995. AU-rich elements: characterization and importance in mRNA degradation. Trends Biochem Sci 20:465–70. [DOI] [PubMed] [Google Scholar]
  14. Cherry JD, Olschowka JA, O’Banion MK. 2014. Neuroinflammation and M2 microglia: the good, the bad, and the inflamed. J Neuroinflammation 11:98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Chiu IM, Morimoto ET, Goodarzi H, Liao JT, O’Keeffe S, Phatnani HP, Muratet M, Carroll MC, Levy S, Tavazoie S and others. 2013. A neurodegeneration-specific gene-expression signature of acutely isolated microglia from an amyotrophic lateral sclerosis mouse model. Cell Rep 4:385–401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Crocker SJ, Frausto RF, Whitton JL, Milner R. 2008. A novel method to establish microglia-free astrocyte cultures: comparison of matrix metalloproteinase expression profiles in pure cultures of astrocytes and microglia. Glia 56:1187–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Cunningham C 2013. Microglia and neurodegeneration: the role of systemic inflammation. Glia 61:71–90. [DOI] [PubMed] [Google Scholar]
  18. David S, Kroner A. 2011. Repertoire of microglial and macrophage responses after spinal cord injury. Nat Rev Neurosci 12:388–99. [DOI] [PubMed] [Google Scholar]
  19. Dean JL, Wait R, Mahtani KR, Sully G, Clark AR, Saklatvala J. 2001. The 3’ untranslated region of tumor necrosis factor alpha mRNA is a target of the mRNA-stabilizing factor HuR. Mol Cell Biol 21:721–730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Dixon DA, Tolley ND, King PH, Nabors LB, McIntyre TM, Zimmerman GA, Prescott SM. 2001. Altered expression of the mRNA stability factor HuR promotes cyclooxygenase-2 expression in colon cancer cells. J Clin Invest 108:1657–1665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Durie D, Lewis SM, Liwak U, Kisilewicz M, Gorospe M, Holcik M. 2011. RNA-binding protein HuR mediates cytoprotection through stimulation of XIAP translation. Oncogene 30:1460–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. El Khoury J, Toft M, Hickman SE, Means TK, Terada K, Geula C, Luster AD. 2007. Ccr2 deficiency impairs microglial accumulation and accelerates progression of Alzheimer-like disease. Nat Med 13:432–438. [DOI] [PubMed] [Google Scholar]
  23. Figuera-Losada M, Rojas C, Slusher BS. 2014. Inhibition of Microglia Activation as a Phenotypic Assay in Early Drug Discovery. J Biomol Screen 19:17–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Filippova N, Yang X, Wang Y, Gillespie GY, Langford C, King PH, Wheeler C, Nabors LB. 2011. The RNA-binding protein HuR promotes glioma growth and treatment resistance. Mol Cancer Res 9:648–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ghosh CC, Ramaswami S, Juvekar A, Vu HY, Galdieri L, Davidson D, Vancurova I. 2010. Gene-specific repression of proinflammatory cytokines in stimulated human macrophages by nuclear IkappaBalpha. J Immunol 185:3685–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Glass CK, Saijo K, Winner B, Marchetto MC, Gage FH. 2010. Mechanisms underlying inflammation in neurodegeneration. Cell 140:918–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hall ED, Oostveen JA, Gurney ME. 1998. Relationship of microglial and astrocytic activation to disease onset and progression in a transgenic model of familial ALS. Glia 23:249–56. [DOI] [PubMed] [Google Scholar]
  28. Hanisch UK. 2002. Microglia as a source and target of cytokines. Glia 40:140–55. [DOI] [PubMed] [Google Scholar]
  29. Henkel J, Beers D, Zhao W, Appel S. 2009. Microglia in ALS: The Good, The Bad, and The Resting. Journal of Neuroimmune Pharmacology 4:389–398. [DOI] [PubMed] [Google Scholar]
  30. Hua B, Tamamori-Adachi M, Luo Y, Tamura K, Morioka M, Fukuda M, Tanaka Y, Kitajima S. 2006. A splice variant of stress response gene ATF3 counteracts NF-kappaB-dependent anti-apoptosis through inhibiting recruitment of CREB-binding protein/p300 coactivator. J Biol Chem 281:1620–9. [DOI] [PubMed] [Google Scholar]
  31. Inoue K 2008. Purinergic systems in microglia. Cell Mol Life Sci 65:3074–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jiang XS, Ni YQ, Liu TJ, Zhang M, Ren H, Jiang R, Huang X, Xu GZ. 2012. CCR2 overexpression promotes the efficient recruitment of retinal microglia in vitro. Mol Vis 18:2982–92. [PMC free article] [PubMed] [Google Scholar]
  33. Katsanou V, Papadaki O, Milatos S, Blackshear PJ, Anderson P, Kollias G, Kontoyiannis DL. 2005. HuR as a Negative Posttranscriptional Modulator in Inflammation. Molecular Cell 19:777–789. [DOI] [PubMed] [Google Scholar]
  34. Kawai T, Lal A, Yang X, Galban S, Mazan-Mamczarz K, Gorospe M. 2006. Translational Control of Cytochrome c by RNA-Binding Proteins TIA-1 and HuR. Mol Cell Biol 26:3295–3307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kawamata T, Akiyama H, Yamada T, McGeer PL. 1992. Immunologic reactions in amyotrophic lateral sclerosis brain and spinal cord tissue. Am J Pathol 140:691–707. [PMC free article] [PubMed] [Google Scholar]
  36. Kettenmann H, Hanisch UK, Noda M, Verkhratsky A. 2011. Physiology of microglia. Physiol Rev 91:461–553. [DOI] [PubMed] [Google Scholar]
  37. King PH, Chen CY. 2014. Role of KSRP in control of type I interferon and cytokine expression. J Interferon Cytokine Res 34:267–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kolaczkowska E, Kubes P. 2013. Neutrophil recruitment and function in health and inflammation. Nat Rev Immunol 13:159–175. [DOI] [PubMed] [Google Scholar]
  39. Lagente V, Le Quement C, Boichot E. 2009. Macrophage metalloelastase (MMP-12) as a target for inflammatory respiratory diseases. Expert Opin Ther Targets 13:287–95. [DOI] [PubMed] [Google Scholar]
  40. Langmead B, Trapnell C, Pop M, Salzberg SL. 2009. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10:R25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Li X, Lin WJ, Chen CY, Si Y, Zhang X, Lu L, Suswam E, Zheng L, King PH. 2012. KSRP: A checkpoint for inflammatory cytokine production in astrocytes. GLIA 60:1773–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Liao G, Wang Z, Lee E, Moreno S, Abuelnasr O, Baudry M, Bi X. 2015. Enhanced expression of matrix metalloproteinase-12 contributes to Npc1 deficiency-induced axonal degeneration. Exp Neurol 269:67–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lorenzl S, Narr S, Angele B, Krell HW, Gregorio J, Kiaei M, Pfister HW, Beal MF. 2006. The matrix metalloproteinases inhibitor Ro 28–2653 [correction of Ro 26–2853] extends survival in transgenic ALS mice. Exp Neurol 200:166–71. [DOI] [PubMed] [Google Scholar]
  44. Lu L, Zheng L, Si Y, Luo W, Dujardin G, Kwan T, Potochick NR, Thompson SR, Schneider DA, King PH. 2014. Hu antigen R (HuR) is a positive regulator of the RNA-binding proteins TDP-43 and FUS/TLS: implications for amyotrophic lateral sclerosis. J Biol Chem 289:31792–804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Mahad DJ, Ransohoff RM. 2003. The role of MCP-1 (CCL2) and CCR2 in multiple sclerosis and experimental autoimmune encephalomyelitis (EAE). Semin Immunol 15:23–32. [DOI] [PubMed] [Google Scholar]
  46. Matsumoto S-i, Kobayashi T, Katoh M, Saito S, Ikeda Y, Kobori M, Masuho Y, Watanabe T. 1998. Expression and Localization of Matrix Metalloproteinase-12 in the Aorta of Cholesterol-Fed Rabbits: Relationship to Lesion Development. The American Journal of Pathology 153:109–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Meisner N-C, Filipowicz W. 2011. Properties of the Regulatory RNA-Binding Protein HuR and its Role in Controlling miRNA Repression In: Großhans H, editor: Springer; New York. p 106–123. [DOI] [PubMed] [Google Scholar]
  48. Meisner N-C, Hintersteiner M, Mueller K, Bauer R, Seifert J-M, Naegeli H-U, Ottl J, Oberer L, Guenat C, Moss S and others. 2007. Identification and mechanistic characterization of low-molecular-weight inhibitors for HuR. Nat Chem Biol 3:508–515. [DOI] [PubMed] [Google Scholar]
  49. Nabors LB, Gillespie GY, Harkins L, King PH. 2001. HuR, an RNA stability factor, is expressed in malignant brain tumors and binds to adenine and uridine-rich elements within the 3’ untranslated regions of cytokine and angiogenic factor mRNAs. Cancer Res 61:2154–2161. [PubMed] [Google Scholar]
  50. Nabors LB, Suswam E, Huang Y, Yang X, Johnson MJ, King PH. 2003. Tumor Necrosis Factor-α Induces Angiogenic Factor Up-Regulation in Malignant Glioma Cells: A Role for RNA Stabilization and HuR. Cancer Res 63:4181–4187. [PubMed] [Google Scholar]
  51. Neumann H, Kotter MR, Franklin RJ. 2009. Debris clearance by microglia: an essential link between degeneration and regeneration. Brain 132:288–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Nguyen MD, Julien JP, Rivest S. 2002. Innate immunity: the missing link in neuroprotection and neurodegeneration? Nat Rev Neurosci 3:216–27. [DOI] [PubMed] [Google Scholar]
  53. Pena-Altamira E, Prati F, Massenzio F, Virgili M, Contestabile A, Bolognesi ML, Monti B. 2015. Changing paradigm to target microglia in neurodegenerative diseases: from anti-inflammatory strategy to active immunomodulation. Expert Opin Ther Targets:1–14. [DOI] [PubMed] [Google Scholar]
  54. Peng SS, Chen C-YA, Xu N, Shyu AB. 1998. RNA stabilization by the AU-rich element binding protein, HuR, an ELAV protein. EMBO J 17:3461–3470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Petrova TV, Akama KT, Van Eldik LJ. 1999. Selective modulation of BV-2 microglial activation by prostaglandin E(2). Differential effects on endotoxin-stimulated cytokine induction. J Biol Chem 274:28823–7. [DOI] [PubMed] [Google Scholar]
  56. Philips T, Robberecht W. 2011. Neuroinflammation in amyotrophic lateral sclerosis: role of glial activation in motor neuron disease. Lancet Neurol 10:253–63. [DOI] [PubMed] [Google Scholar]
  57. Philips T, Rothstein JD. 2014. Glial cells in amyotrophic lateral sclerosis. Exp Neurol 262 Pt B:111–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Power C, Henry S, Del Bigio MR, Larsen PH, Corbett D, Imai Y, Yong VW, Peeling J. 2003. Intracerebral hemorrhage induces macrophage activation and matrix metalloproteinases. Ann Neurol 53:731–42. [DOI] [PubMed] [Google Scholar]
  59. Ransohoff R, Schafer D, Vincent A, Blachère N, Bar-Or A. 2015. Neuroinflammation: Ways in Which the Immune System Affects the Brain. Neurotherapeutics:1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Rhee WJ, Ni C-W, Zheng Z, Chang K, Jo H, Bao G. 2010. HuR regulates the expression of stress-sensitive genes and mediates inflammatory response in human umbilical vein endothelial cells. Proc Natl Acad Sci USA 107:6858–6863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Rodriguez-Pascual F, Hausding M, Ihrig-Biedert I, Furneaux H, Levy AP, Forstermann U, Kleinert H. 2000. Complex contribution of the 3’-untranslated region to the expressional regulation of the human inducible nitric-oxide synthase gene. Involvement of the RNA-binding protein HuR. J Biol Chem 275:26040–26049. [DOI] [PubMed] [Google Scholar]
  62. Semple BD, Kossmann T, Morganti-Kossmann MC. 2010. Role of chemokines in CNS health and pathology: a focus on the CCL2/CCR2 and CXCL8/CXCR2 networks. J Cereb Blood Flow Metab 30:459–473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Sen R, Smale ST. 2010. Selectivity of the NF-{kappa}B response. Cold Spring Harb Perspect Biol 2:a000257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Shipley JM, Wesselschmidt RL, Kobayashi DK, Ley TJ, Shapiro SD. 1996. Metalloelastase is required for macrophage-mediated proteolysis and matrix invasion in mice. Proc Natl Acad Sci U S A 93:3942–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Shiryaev SA, Savinov AY, Cieplak P, Ratnikov BI, Motamedchaboki K, Smith JW, Strongin AY. 2009. Matrix Metalloproteinase Proteolysis of the Myelin Basic Protein Isoforms Is a Source of Immunogenic Peptides in Autoimmune Multiple Sclerosis. PLoS ONE 4:e4952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Shyu A-B, Greenberg ME, Belasco JG. 1989. The c-fos transcript is targeted for rapid decay by two distinct mRNA degradation pathways. Genes Dev 3:60–72. [DOI] [PubMed] [Google Scholar]
  67. Simpson J, Rezaie P, Newcombe J, Cuzner ML, Male D, Woodroofe MN. 2000. Expression of the β-chemokine receptors CCR2, CCR3 and CCR5 in multiple sclerosis central nervous system tissue. J Neuroimmunol 108:192–200. [DOI] [PubMed] [Google Scholar]
  68. Slater K 2001. Cytotoxicity tests for high-throughput drug discovery. Curr Opin Biotechnol 12:70–4. [DOI] [PubMed] [Google Scholar]
  69. Srikantan S, Gorospe M. 2012. HuR function in disease. Front Biosci (Landmark Ed) 17:189–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Streit WJ. 2002. Microglia as neuroprotective, immunocompetent cells of the CNS. Glia 40:133–139. [DOI] [PubMed] [Google Scholar]
  71. Suswam EA, Li YY, Mahtani H, King PH. 2005. Novel DNA-binding properties of the RNA-binding protein TIAR. Nucleic Acids Res 33:4507–4518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Taylor PR, Roy S, Leal SM Jr., Sun Y, Howell SJ, Cobb BA, Li X, Pearlman E. 2014. Activation of neutrophils by autocrine IL-17A-IL-17RC interactions during fungal infection is regulated by IL-6, IL-23, RORgammat and dectin-2. Nat Immunol 15:143–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Town T, Nikolic V, Tan J. 2005. The microglial “activation” continuum: from innate to adaptive responses. J Neuroinflammation 2:24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Trapnell C, Pachter L, Salzberg SL. 2009. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25:1105–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL, Pachter L. 2012. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc 7:562–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L. 2010. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28:511–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Turner MR, Cagnin A, Turkheimer FE, Miller CC, Shaw CE, Brooks DJ, Leigh PN, Banati RB. 2004. Evidence of widespread cerebral microglial activation in amyotrophic lateral sclerosis: an [11C](R)-PK11195 positron emission tomography study. Neurobiol Dis 15:601–9. [DOI] [PubMed] [Google Scholar]
  78. Vos CM, van Haastert ES, de Groot CJ, van der Valk P, de Vries HE. 2003. Matrix metalloproteinase-12 is expressed in phagocytotic macrophages in active multiple sclerosis lesions. J Neuroimmunol 138:106–14. [DOI] [PubMed] [Google Scholar]
  79. Wake H, Moorhouse AJ, Miyamoto A, Nabekura J. 2013. Microglia: actively surveying and shaping neuronal circuit structure and function. Trends Neurosci 36:209–17. [DOI] [PubMed] [Google Scholar]
  80. Wells JE, Rice TK, Nuttall RK, Edwards DR, Zekki H, Rivest S, Yong VW. 2003. An adverse role for matrix metalloproteinase 12 after spinal cord injury in mice. J Neurosci 23:10107–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Weydt P, Yuen EC, Ransom BR, Möller T. 2004. Increased cytotoxic potential of microglia from ALS-transgenic mice. Glia 48:179–182. [DOI] [PubMed] [Google Scholar]
  82. Xia MQ, Qin SX, Wu LJ, Mackay CR, Hyman BT. 1998. Immunohistochemical study of the beta-chemokine receptors CCR3 and CCR5 and their ligands in normal and Alzheimer’s disease brains. Am J Pathol 153:31–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  83. Zhao W, Beers DR, Appel SH. 2013. Immune-mediated Mechanisms in the Pathoprogression of Amyotrophic Lateral Sclerosis. J Neuroimmune Pharmacol 8:888–899. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1672569-supplement-1.pdf (3MB, pdf)

RESOURCES